[Characteristics of serum bile acids among healthy children in Zhejiang province].

Objective: To characterize the serum bile acid profiles of healthy children in Zhejiang Province. Methods: A cross-sectional study was conducted on 245 healthy children who underwent imaging and laboratory biochemical tests during routine physical examinations at the Children's Hospital of Zhejiang University School of Medicine from January 2020 to July 2022. Overnight fasting venous blood samples were collected, and the concentrations of 18 individual bile acids in the serum were accurately quantitated using tandem mass spectrometry. The concentration difference of bile acid were compared between different genders and to explore the correlation between age and bile acid levels. Used the Mann-Whitney U test for intergroup comparison and Spearman test to correlation analysis. Results: A total of 245 health children with a age of 10 (8, 12) years including 125 boys and 120 girls. There were no significant differences in levels of total bile acids, primary and secondary bile acids, free and conjugated bile acids between the two gender groups (all P>0.05). The serum concentrations of ursodeoxycholic acid and glycoursodeoxycholic acid in girls were significantly higher than those in boys (199.0 (66.9, 276.5) vs. 154.7 (49.3, 205.0) nmol/L, 274.0 (64.8, 308.0) vs. 181.0 (43.8, 209.3) nmol/L, Z=2.06, 2.71, both P<0.05). The serum taurolithocholic acid in both boys and girls were positively correlated with age (r=0.31, 0.32, both P<0.05). The serum chenodeoxycholic acid and glycochenodeoxycholic acid in the boys group were positively correlated with age (r=0.20, 0.23, both P<0.05), whereas the serum tauroursodeoxycholic acid in the girls group was negatively correlated with age (r=-0.27, P<0.05), and the serum cholic acid was positively correlated with age (r=0.34, P<0.05). Conclusions: The total bile acid levels are relatively stable in healthy children in Zhejiang province. However, individual bile acids showed gender differences and were correlated with age.

[Clinical analysis of eosinophilic gastroenteritis in 71 children].

Objective: To summarize the clinical data including manifestations, diagnosis, treatment and prognosis of eosinophilic gastroenteritis (EGE) in children. Methods: A retrospective analysis was performed in 71 patients with pathologically proven EGE at Beijing Children's Hospital Affiliated to Capital Medical University from January 2008 to January 2017. Their clinical manifestations, laboratory and imaging examinations, endoscopic findings, histopathological examinations, and treatment were collected and analyzed. Results: Among 71 EGE cases, 47 (66%) cases were male and 24 (34%) cases were female, and the median age was 9.2 (0.2-16.5) years old. The main clinical manifestations included abdominal pain (76%, 54/71), vomiting (68%, 48/71), anorexia (54%, 38/71), weight loss (38%, 27/71), and diarrhea (37%, 26/71). There were 27 cases (38%) with a history of allergic diseases or family history. The median absolute value of eosinophil in peripheral blood of the 71 patients was 0.4 (0-36.8)×10(9)/L, and 27 cases (38%) showed an increase in eosinophil counts. Serum IgE was measured in 52 patients (104.3 (3.4- 3 000.0)×10(3) U/L), and 30 patients (58%) showed an increase in serum IgE. A large number of eosinophils ((41.0±8.5)/HP) were found in 3 patients' ascites. The endoscopic examination of upper gastrointestinal tract revealed hyperemic edema in 62 cases (87%), plaque in 44 cases (62%), erosion in 17 cases (24%) and ulceration in 16 cases (23%). Histopathologically, in 8 cases (11%) the disease involved both stomach and duodeneum, in 21 cases (30%) involved stomach only, and in 37 cases (52%) involved duodeneum only. In addition, in 6 cases (8%) the disease involved esophagus and in 10 cases (14%) involved colorectum. Microscopically, eosinophil counts averaged 67/HP, 33/HP, 40/HP and 38/HP in esophageal, gastric, duodenal and colorectal mucosa respectively. A total of 34 cases were treated with glucocorticoid, and all these patients had alleviation of symptoms, which occurred within 14.9 days on average, but EGE recurred in 11 cases (32%). Conclusions: The clinical symptoms and endoscopic findings of EGE are diverse and nonspecific. Histopathological examination of gastrointestinal mucosa is particularly important for the diagnosis. Glucocorticoid treatment is effective, but the patients with EGE are prone to relapse.

[Clinical analysis of intestinal lymphangiectasia in 47 children].

Objective: To analyze the clinical manifestations, diagnosis, treatment and prognosis of intestinal lymphangiectasia (IL) in children in order to improve the skills of diagnosis and treatment of IL. Method: Clinical manifestations, laboratory findings, gastroscopic findings, histopathological examinations and lymphatic radionuclide imaging assessments were analyzed retrospectively among 47 IL patients who were hospitalized in the Gastroenterology Department of Beijing Children's Hospital Affiliated to Capital Medical University from June 2007 to December 2015. All patients were followed up by telephone. According to the various causes, the patients were divided into the primary intestinal lymphangiectasia (PIL) group and secondary IL group, and their clinical manifestations were compared by t test, Rank sum test or Chi-square test. Result: In 47 IL patients, there were 38 children (81%) younger than 3 years old. There were 43 PIL patients (91%) and 4 secondary IL patients (9%). Between PIL and secondary IL, there were statistical differences in serum albumin (t=-3.950, P<0.005) , globulin(t=-2.850, P=0.007), age of onset(U=27.000, P=0.024), age at diagnosis(U=29.000, P=0.030) and course of disease(U=26.500, P=0.023), whereas there were no statistical differences in lymphocyte count, IgG, lymphatic radionuclide imaging, histopathology and gender(all P>0.05). Edema (44 cases, 94%), diarrhea (42 cases, 89%), accompanied with infection (35 cases, 74%) and ascites (30 cases, 64%) were the main clinical manifestations. In 47 IL patients, 45 patients were done gastroscopy and histopathological examinations, and there were 31 patients' histopathological examinations(69%) were positive. Forty patients were done lymphatic radionuclide imaging, and there was evidence of protein losing from gut via lymphatic radionuclide imaging in 39 patients(98%). Among 47 patients, 35 patients (74%) were followed up, 32 patients had good prognosis, 2 patient failed to show evidence of improvement, 1 patient died and no patient experienced a relapse till the end of the follow-up. In 35 patients, 28 patients were treated with medium chain triglycerides (MCT) dietary therapy, 26 patients showed improvement in symptoms, and 2 patients had no improvement. Among 35 patients with follow-up, there were 6 patients received surgical treatment, and their symptoms were improved. Conclusion: PIL are the majority of IL in children younger than 3 years old. The main clinical manifestations are edema, diarrhea, accompanied with infection and ascites. For the patients without the evidence of lymphangiectasia from duodenum histopathological examination, further consideration of lymphatic radionuclide imaging, clinical manifestations, and laboratory studies are needed to make a final diagnosis. MCT dietary therapy is the cornerstone of IL medical management.

MiR-146a inhibits proliferation and induces apoptosis in murine osteoblastic MC3T3-E1 by regulating Bcl2.

OBJECTIVE: The dysregulation of proliferation and apoptosis plays a significant role in the pathogenesis of hormone-induced osteonecrosis of femoral head (ONFH). The research aimed to explore the regulatory role of miR-146a in dexamethasone (DEX)-induced proliferation and apoptosis change in MC3T3-E1 cells from murine osteoblastic. MATERIAL AND METHODS: In this study, MC3T3-E1 was co-cultured with 10-7 DEX for 6 h, then RT-PCR was employed to test the expression level of miR-146a and Bcl2. CCK8 assay and flow cytometry were adopted to verify miR-146a could regulate proliferation and apoptosis. After transfected MC3T3-E1 with mimics and inhibitor, RT-PCR and Western blot was used to detect Bcl2 expression level. RESULTS: In DEX treated MC3T3-E1 cells showed higher MiR-146a expression level and lower Bcl2 expression level. MiR-146a could inhibit proliferation and promotes apoptosis in murine osteoblastic MC3T3-E1 cells. Additionally, Bcl2 gene is regulated by MiR-146a. CONCLUSIONS: The MiR-146a expression level increased, while Bcl2 has low expression level in dexamethasone treated MC3T3-E1 cells. MiR-146a regulates proliferation and apoptosis of mouse bone cells. The low expression level of Bcl2 in DEX treated MC3T3-E1 cells is caused by increased MiR-146a level.

[Clinical analysis of 11 children with pancreatic cystic fibrosis].

Objective: To increase the recognition of pancreatic cystic fibrosis (PCF) in children and facilitate diagnosing and treatment of this rare entity. Method: This is a retrospective analysis of children who presented to Beijing Children's Hospital affiliated to Capital Medical University from January 2010 to December 2015. We describe their clinical features, laboratory testing and management. Result: Eleven children were diagnosed with PCF by genetic testing or sweat chloride test during these 5 years, including 4 boys and 7 girls. Their age ranged from 0.5-14.3 (mean 9.0±3.9) years. Family history was positive in 3 children. Significant clinical findings on presentation were: malnutrition 6, including 2 cases of mild, moderate and severe malnutrition each; diarrhea 4 (yellow mushy or watery stool with frequency ranging from 2-5 times a day), including 1 case of acute diarrhea and 3 of chronic diarrhea, 3 of them had steatorrhea; abdominal pain 3. All of them had pancreatic lesions shown by abdominal ultrasound. Blood tests showed 6 cases had elevated serum amylase and lipase. The main treatment was pancreatic replacement therapy and nutritional support. Conclusion: PCF is rare in children. Malnutrition, diarrhea and abdominal pain are the main clinical manifestations. Treatment is mostly pancreatic enzymes replacement and supportive care.

A genome-wide scan reveals candidate susceptibility loci for pig hernias in an intercross between White Duroc and Erhualian.

Pig scrotal/inguinal and umbilical hernias are the most prevalent congenital disorders in pigs and often cause animal welfare problems and economic loss. To identify susceptibility loci for these traits, a genome-wide scan with 194 microsatellite markers covering the pig genome was performed in a White Duroc x Erhualian resource population with 23 scrotal/inguinal F(2) animals, 50 umbilical F(2) animals, and their unaffected siblings. A sex-average linkage map with a total length of 2,350.3 cM and an average marker interval of 12.84 cM was constructed. Both nonparametric genome-wide linkage (NPL) analysis and transmission disequilibrium test (TDT) were implemented to detect closely linked markers. The NPL analysis revealed 11 chromosomal regions on SSC1, 2, 3, 6, 7, 8, 10, and 11 for umbilical hernia and 5 regions on SSC2, 4, 8, 13, and 16 for scrotal/inguinal hernia, whereas the TDT test identified susceptibility loci for umbilical hernia on SSC1, 2, 4, 7, 10, 13, 14, and 15 and for scrotal/inguinal hernias on SSC2, 8, 10, and 18. The most promising loci were SWR1928 on SSC7 and SW830 on SSC10 for umbilical hernia, and SW933 on SSC8 for scrotal hernia, which were consistently detected by both NPL and TDT. Several previously reported chromosomal regions for scrotal/inguinal hernia were confirmed, and new evidence for linkage with this pig defect was found. Moreover, susceptibility loci for pig umbilical hernia were detected for the first time.

[Changes of the certain functions of mitochondria induced by acute exercise and the protective effect of bilirubin].

AIM: To study the mechanism of fatigue induced by acute exercise and the protective effect of bilirubin. METHODS: 41 Wistar male rats were divided into five groups: control group, exercise group, exercise recovery group, bilirubin-treated exercise group and bilirubin-treated exercise recovery group. The rats were administrated with 1 micromol/kg body weight of bilirubin or saline once every day for 4 weeks, after swimming with load for 2 h, all of the rats were killed and several index were determined. RESULTS: Acute exercise could induce the increase of Ca2+ content in cytoplasm and mitochondria of gastrocnemius, Ca2+ content increased continuously after 12 hours in mitochondria. The Ca2(+)-Mg2(+)-ATPase activity decreased obviously after acute exercise, and the activity increased again after 12 hours. Bilirubin could inhibit these changes. The change of Mg2(+)-ATPase activity in mitochondria had little difference between bilirubin-treated and untreated groups, both of which were lower than that of control group, but the recovery was faster in the bilirubin-treated group than that in untreated group. CONCLUSION: Bilirubin could protect muscle fibers from injury induced by acute exercise and delay the development of fatigue and promote the recovery through inhibiting the disturbance of the certain function of mitochondria.

Mice with gene disruption of both endothelial and neuronal nitric oxide synthase exhibit insulin resistance.

Studies from our laboratory using acute pharmacologic blockade of nitric oxide synthase (NOS) activity have suggested that nitric oxide (NO) has an important role in regulating carbohydrate metabolism. We now report on insulin sensitivity in mice with targeted disruptions in endothelial NOS (eNOS) and neuronal NOS (nNOS) genes compared with their wild-type (WT) counterparts. Mice underwent hyperinsulinemic-euglycemic clamp studies after a 24-h fast, during an insulin infusion of 20 mU x kg(-1) x min(-1). Glucose levels were measured at baseline and every 10 min during the clamp. Insulin levels were measured at baseline and at the end of the clamp study. Glucose infusion rates (GIRs) during the last 30 min of the clamp study were in a steady state. Tritiated glucose infusion was used to measure rates of endogenous glucose output (EGO) both at baseline and during steady-state euglycemia. Glucose disposal rates (GDRs) were computed from the GIR and EGO. Fasting and steady-state glucose and insulin levels were comparable in the 3 groups of mice. No differences in fasting EGO were noted between the groups. GIR was significantly reduced (37%, P = 0.001) in the eNOS knockout (KO) mice compared with the WT mice, with values for the nNOS mice being intermediate. EGO was completely suppressed in the nNOS and WT mice during insulin infusion, but not in the eNOS mice. Even so, the eNOS mice displayed significantly reduced whole-body GDRs compared with those of the WT mice (82.67+/-10.77 vs. 103.67+/-3.47 mg x kg(-1) x min(-1), P = 0.03). eNOS KO mice are insulin resistant at the level of the liver and peripheral tissues, whereas the nNOS KO mice are insulin resistant only in the latter. These data indicate that NO plays a role in modulating insulin sensitivity and carbohydrate metabolism and that the eNOS isoform may play a dominant role relative to nNOS.

Dose-response relationship of insulin to glucose fluxes in the awake and unrestrained mouse.

The purpose of the study was to use the hyperinsulinemic-euglycemic clamp technique to generate insulin dose-response curves for insulin suppression of endogenous glucose output (EGO) and stimulation of the glucose disposal rate (GDR) in conscious unstressed mice. Five groups of male ICR (Institute for Cancer Research) mice were studied (N = 43). The animals underwent surgery for implantation of a jugular vein catheter 2 to 3 days before the clamp and were fasted 6 hours before the study. Each group was clamped at a different insulin infusion rate of 0, 2.5, 10, or 20 mU/kg/min. 3H-3-glucose was infused for measurement of the glucose turnover rate (rate of appearance [Ra]). Blood samples were collected by milking a severed tail-tip. EGO was calculated as the difference between the Ra and glucose infusion rate (GIR), and the glucose clearance rate (GCR) as the GDR divided by the plasma glucose concentration. From the curves generated, half-maximal EGO and GCR were obtained at a plasma insulin concentration of 20 to 30 microU/mL, which was achieved at an insulin infusion rate of about 4 to 5 mU/kg/min. Maximal suppression of EGO and stimulation of the GCR occurred at an insulin infusion rate of 10 mU/kg/min. The establishment of normative curves for insulin-stimulated glucose metabolism in conscious mice facilitates the evaluation of glucose metabolism in a variety of mouse models of insulin resistance.

Glucosamine infusion in rats rapidly impairs insulin stimulation of phosphoinositide 3-kinase but does not alter activation of Akt/protein kinase B in skeletal muscle.

Glucosamine, a metabolite of glucose via the hexosamine biosynthetic pathway, potently induces insulin resistance in skeletal muscle by impairing insulin-induced GLUT4 translocation to the plasma membrane. Activation of phosphoinositide (PI) 3-kinase is necessary for insulin-stimulated GLUT4 translocation, and the serine/threonine kinase Akt/protein kinase B (PKB) is a downstream mediator of some actions of PI 3-kinase. To determine whether glucosamine-induced insulin resistance could be due to impaired signaling, we measured insulin receptor substrate (IRS)-1 and insulin receptor tyrosine phosphorylation; PI 3-kinase activity associated with IRS-1, IRS-2, and phosphotyrosine; and Akt activity and phosphorylation in skeletal muscle of rats infused for 2 h with glucosamine (6.0 mg x kg(-1) x min(-1)) or saline. Euglycemic-hyperinsulinemic clamp studies (12 mU x kg(-1) x min(-1) insulin) in awake rats showed that glucosamine infusion resulted in rapid induction of insulin resistance, with a 33% decrease in glucose infusion rate (P < 0.01). Tissues were harvested after saline alone (basal), 1 min after an insulin bolus (10 U/kg), or after 2 h of insulin clamp in saline- and glucosamine-infused rats. After 1 min of insulin stimulation, phosphorylation of IRS-1 and insulin receptor increased 6- to 8-fold in saline-infused rats and 7- to 10-fold in glucosamine-infused rats. In saline-infused rats, 1 min of insulin stimulation increased PI 3-kinase activity associated with IRS-1, IRS-2, or phosphotyrosine 7.6-, 6.4-, and 10-fold, respectively. In glucosamine-infused rats treated for 1 min with insulin, PI 3-kinase activity associated with IRS-1 was reduced 28% (P < 0.01) and that associated with phosphotyrosine was reduced 43% (P < 0.01). Insulin for 1 min stimulated Akt/PKB activity approximately 5-fold in both saline- and glucosamine-infused rats; insulin-induced hyperphosphorylation of Akt/PKB was not different between groups. Glucosamine infusion alone had no effect on tyrosine phosphorylation of the insulin receptor or IRS-1 or on stimulation of PI 3-kinase or Akt/PKB activity. However, 2 h of insulin clamp reduced PI 3-kinase activity associated with IRS-1, IRS-2, or phosphotyrosine to <30% of that seen with 1 min of insulin. No effect of glucosamine was seen on these signaling events when compared with 2 h of insulin clamp without glucosamine. Our data show that 1) glucosamine infusion in rats is associated with an impairment in the early activation of PI 3-kinase by insulin in skeletal muscle, 2) this insulin-resistant state does not involve alterations in the activation of Akt/PKB, and 3) prolonged insulin infusion under clamp conditions results in a blunting of the PI 3-kinase response to insulin.

Central nervous system nitric oxide synthase activity regulates insulin secretion and insulin action.

Systemic inhibition of nitric oxide synthase (NOS) with NG-monomethyl-L-arginine (L-NMMA) causes acute insulin resistance (IR), but the mechanism is unknown. We tested whether L-NMMA-induced IR occurs via NOS blockade in the central nervous system (CNS). Six groups of Sprague-Dawley rats were studied after chronic implantation of an intracerebroventricular (ICV) catheter into the lateral ventricle and catheters into the carotid artery and jugular vein. Animals were studied after overnight food deprivation, awake, unrestrained, and unstressed; all ICV infusion of L-NMMA or D-NMMA (control) were performed with artificial cerebrospinal fluid. ICV administration of L-NMMA resulted in a 30% rise in the basal glucose level after 2 h, while ICV D-NMMA had no effect on glucose levels. Insulin, epinephrine, and norepinephrine levels were unchanged from baseline in both groups. Tracer (3H-3-glucose)-determined glucose disposal rates during 2 h euglycemic hyperinsulinemic (300 microU/ml) clamps performed after ICV administration of L-NMMA were reduced by 22% compared with D-NMMA. Insulin secretory responses to a hyperglycemic clamp and to a superimposed arginine bolus were reduced by 28% in L-NMMA-infused rats compared with D-NMMA. In conclusion, ICV administration of L-NMMA causes hyperglycemia via the induction of defects in insulin secretion and insulin action, thus recapitulating abnormalities observed in type 2 diabetes. The data suggest the novel concept that central NOS-dependent pathways may control peripheral insulin action and secretion. This control is not likely to be mediated via adrenergic mechanisms and could occur via nonadrenergic, noncholinergic nitrergic neural and/or endocrine pathways. These data support previously published data suggesting that CNS mechanisms may be involved in the pathogenesis of some forms of insulin resistance and type 2 diabetes independent of adiposity.

The effect of glucose and glucagon-like peptide-1 stimulation on insulin release in the perfused pancreas in a non-insulin-dependent diabetes mellitus animal model.

This study was designed to investigate the effect of glucogon-like peptide-1 (GLP-1) on pancreatic beta-cell function in normal, Zucker diabetic fatty (ZDF) rats, a model for non-insulin-dependent diabetes mellitus (NIDDM or type II diabetes) and their heterozygous siblings. Pancreas perfusion and enzyme-linked immunosorbent assay (ELISA) were used to detect the changes in insulin release under fasting and hyperglycemic conditions and following stimulation with GLP-1. Animals from the ZDF/Gmi-fa rats (ZDF) were grouped according to age, sex, and phenotype (obese or lean), and compared with LA lean rats. Glucose stimulation (10 mmol/L) in obese rats showed repressed response in insulin release. Glucose plus GLP-1 stimulation caused increased insulin release in all groups. The degree of this response differed between groups: lean > obese; young > adult; female > male. The LA lean control group was most sensitive, while the ZDF overtly diabetic group had the lowest response. In addition, the pulsatile pattern of insulin secretion was suppressed in ZDF rats, especially in obese groups. These results support the hypothesis that GLP-1 can effectively stimulate insulin secretion. Insulin release was defective in ZDF obese rats and could be partially restored with GLP-1. ZDF lean rats also showed suppression of beta-cell function and there was a difference in beta-cell function related to sex in ZDF strain. This study documents the efficacy of GLP-1 to stimulate insulin release and contributes to our understanding of the pathophysiological mechanisms underlying NIDDM.

Removal of a dimeric form of surfactant protein C from mouse lungs: its acceleration by reduction.

Clearance of hydrophobic surfactant-associated protein C (SP-C) and its dimeric form ([SP-C]2) was investigated. SP-C and [SP-C]2 obtained from proteinosis patients were fluorescently labeled and were instilled into mouse lungs as lipid-protein complexes. [SP-C]2 was removed more slowly than SP-C, with apparent half-lives of 30 and 18 h, respectively. A significant amount of [SP-C]2 was removed as SP-C, and the conversion rate was 0.22 micrograms.h-1.mouse-1. By correcting the removal as SP-C, we obtained 38 h for a possible half-life of [SP-C]2. Conversion from SP-C to [SP-C]2 seemed very slow. Decrease in glutathione (GSH) in the lung inhibited the conversion of [SP-C]2 to SP-C and GSH-treatment of liposomes accelerated clearance of [SP-C]2. These results suggest that the removal of [SP-C]2 from lung is accelerated by reduction and that GSH acts as a reducing agent in the lung.

Effects of proteinosis surfactant proteins on the viability of rat alveolar macrophages.

Pulmonary alveolar proteinosis (PAP) is a disease of unknown etiology that is characterized by the accumulation of protein- and lipid-rich insoluble material in alveoli and terminal bronchioles of the lung. Alveolar macrophages (AM) in PAP are reportedly extremely large and have low viability. We investigated substances in lavaged lung material from patients with PAP that induced these cellular changes in rat AM. Rat AM were incubated for various periods with liposomes prepared from lipids and isolated hydrophobic surfactant apoproteins, and cell size, viability, and lactate dehydrogenase release were determined. Of the hydrophobic apoproteins, a dimeric form of surfactant-associated protein-C ([SP-C]2) had the most marked effects. In addition, [SP-C]2 induced increased superoxide anion release at an early phase (6 to 12 h) and an increase in glutathione content at 24 h of incubation. At 3 d after incubation, cellular glutathione and adenosine triphosphate (ATP) content were significantly decreased in cells treated with [SP-C]2, [SP-C]2 was presumed to cause early cell death through increased formation of superoxide anion and the subsequent derangement of cellular metabolism. [SP-C]2 was not removed from cells, and SP-B and SP-C were removed at slower rates than lipids. The changes in macrophages induced by [SP-C]2 may contribute to establishing PAP.

Surfactant protein A accumulating in the alveoli of patients with pulmonary alveolar proteinosis: oligomeric structure and interaction with lipids.

Pulmonary alveolar proteinosis (PAP) is a diffuse lung disease of unknown etiology in which the alveoli and terminal bronchioles of the lung fill with large amounts of surfactant-rich lipoproteinaceous materials. Its major pathologic manifestations are a small number of normal tubular myelin structures and an unusual abundance of multilamellated structures. Since surfactant protein A (SP-A) plays an important role in surfactant phospholipid homeostasis, we investigated the structural features of SP-A oligomers (alveolar proteinosis protein, APP) accumulating in the alveoli of individuals with PAP, and examined the abilities of APP to interact with lipids. Analysis of APP by Bio Gel A15m column chromatography revealed that it was composed of two protein peaks, one of which (APP-I) eluted at the position near that of blue dextran whereas the other (APP-II) eluted far behind blue dextran but ahead of thyroglobulin. These populations of APP showed almost identical amino acid compositions. Electron microscopic observations of APP molecules using the rotary shadow technique revealed that APP-II was observed as hexameric particles, presumably consisting mainly of octadecamers whose diameter was approximately 30 nm. The population seen for APP-II was similar to that seen for SP-A from healthy individuals. In contrast, APP-I was observed as multimerized larger aggregates whose diameter appeared to be about 70 to 90 nm. Both APP-I and APP-II retained the abilities to bind dipalmitoylphosphatidylcholine (DPPC). They also induced phospholipid vesicle aggregation in a concentration-dependent manner. The maximal turbidity for light scattering induced by APP-I and APP-II was almost equivalent when analyzed as a function of molar concentration. In vitro reconstitution experiments with porcine surfactant protein B (SP-B) and phospholipids revealed that the multilamellated membranes in structures formed from APP-I consisted of several layers of doubled unit membranes. APP-I failed to form tubular myelin structures. In contrast, APP-II formed well-formed lattice structures seen in tubular myelin. From these data we conclude that there exists an abnormal multimerized form of SP-A oligomer in the alveoli of patients with PAP, and that this unusual subpopulation of SP-A oligomer exhibits abnormal function on phospholipid membrane organization.

Analysis of fused-membrane structures in bronchoalveolar lavage fluid from patients with alveolar proteinosis.

Alveolar macrophages of patients with alveolar proteinosis, obtained by lung lavage, contain a large number of intracytoplasmic inclusions with mutlilamellar membranous structures, called fused-membrane structures, having a periodicity of 4.7 nm. To analyze the composition of these structures, we concentrated them from the bronchoalveolar lavage fluid of patients by a combination of sucrose density-gradient ultracentrifugation and enzyme hydrolysis using proteinase K. Fused-membrane structures were most numerous (26.2%) in a fraction obtained at the interface between 1.0 and 1.1 M sucrose solutions separated after proteinase K hydrolysis. This fraction was rich in acidic phospholipids. Hydrophobic surfactant apoproteins constituted about 77% of the proteins in this fraction, and a remarkable increase in the content of surfactant-associated protein C (SP-C) was found. The phospholipid-to-protein ratio was 0.25:1. Based on these results, we tried to reconstitute fused-membrane structures from purified lipids and hydrophobic proteins isolated from the patients' lavage fluid or from pig lungs. Only in the presence of both phospholipids and SP-C were similar multilamellar structures, having a periodicity of 4.3 to 4.5 nm, formed. These results suggest that fused-membrane structures have a close relationship to a hydrophobic surfactant-associated protein, SP-C, which accumulates in alveolar macrophages, possibly by incomplete digestion.